Re: pterosaur take-off analog

[Mike Habib <habib@jhmi.edu>]

On Oct 5, 2009, at 11:07 AM, David Peters wrote:

> 1. Knowing when and where to land
> 2. Knowing to land in trees tops when the ground was "busy"
> 3. Knowing to land on deserted islands or flotsam when trees weren't  
> around.
> 4. Knowing when the breezy parts of the day occurred to land and  
> take-off.
> 5. Landing somewhere with a ten-foot drop off nearby (probably a  
> traditional site)
> 6. Climbing up a nearby tree or vertical surface to launch (not  
> exactly like WAIR, but maybe)
> 7. Or simply launching in some sort of springy fashion if all else  
> failed.

It is quite reasonable to presume that numbers 1-6 would be used, as  
appropriate, by various species.  However, all living powered flyers  
are capable of number 7, as well, and what is more, their primary  
launch mode is generally some version thereof.  In other words, living  
birds and bats have structural adaptations to launch is your "some  
sort of springy fashion", and this appears to be the mode that drives  
their launch-related morphological evolution.  Assuming that the same  
is true for pterosaurs, their morphology should reflect their primary  
launch mode, which would be the one used without special conditions.   
Wind and elevation then helps from there.


> While the recent animation on pterosaur arm launches is an  
> interesting look at engineering and dynamics, it depends on a worst- 
> case scenario of a flat surface and no breeze.

Of course; it does little good to layer on special conditions and then  
presume that the result is a typical launch that properly reflects the  
adaptations of the animals in question.

> Given any other scenario, other solutions may be more tenable given  
> the lesson of the fledglings.

I'm a bit intrigued by your use of "tenable" here.  What I think you  
are saying is that other launch solutions may be vaguely plausible,  
given special conditions, and presuming that such conditions were  
always available.  That's not the same as being "more tenable".   
Consider: the quad launch model matches with the known quad stance  
(trackway evidence), matches the structural evidence (strength of  
bones in bending), explains the size differential between the largest  
pterosaurs and all other flying animals, and explains the observed  
patterns of maximum size within Pterosauria.  Biped launch models fail  
on all these accounts, and furthermore have the issue of trailing edge  
flutter and angle of attack.  In fact, right off the starting line,  
the quad launch model is the more parsimonious, as it entails a quad- 
walking animal also quad launching.  A biped model presumes a shift in  
gait (at a mechanical disadvantage, no less) from quad to biped. So  
really, all I, Jim, and others have done is gather some quantitative  
evidence to support what should be the null model.  No evidence has  
yet been accrued to refute the null hypothesis and/or support an  
alternative.


> There's still the problem of no clear metacarpal IV impressions,  
> only digits I-III. So there may not be a lock and recoil mechanism.

Well, I think there is a huge MCIV impression, namely the middle of  
almost every manus track.  Several track workers have agreed with me  
here.  However, let's presume that the tracks really are just 1-3  
(which would be odd for a number of reasons, more on that some other  
time); as long as MCIV can be pressed against the substrate during  
preload, the lock and release would work.  Going even further, and  
presuming that for some reason MCIV just cannot touch the substrate,  
and that things like Quetz somehow levitate to support themselves on  
digits 1-3, the lock and release is not required for launch: it just  
makes it faster.  In fact, the low-power version of the launch I  
calculated used only counter-motion preload (and not even much of  
that); it had no elastic storage at all.

> And the problem of what to do with that giant wing finger rotating  
> straight down at the moment of launch is still vexing.

If it did so, then it might be vexing.  We've discussed this  
previously: the wing finger does not rotate straight down; it stays  
mostly closed until the first upstroke (it opens about 10 degrees  
during the catch and release trigger).  Note, for example, that we did  
check for clearance with the ground in the digital model, and the  
animation used published joint limits, so we did not have to do  
anything fancy for clearance (clearance is improved by quad launching,  
of course, because there is more height gain during the leap).

Cheers,

--Mike



Michael Habib
Assistant Professor of Biology
Chatham University
Woodland Road, Pittsburgh PA  15232
Buhl Hall, Room 226A
mhabib@chatham.edu
(443) 280-0181